Forum for Science, Industry and Business

Sponsored by:     3M 
Search our Site:

 

System automatically retouches cellphone images in real-time

03.08.2017

New system can apply a range of styles in real-time, so that the viewfinder displays the enhanced image.

The data captured by today's digital cameras is often treated as the raw material of a final image. Before uploading pictures to social networking sites, even casual cellphone photographers might spend a minute or two balancing color and tuning contrast, with one of the many popular image-processing programs now available.


Researchers describe a new system that can automatically retouch images in the style of a professional photographer. It can run on a cellphone and it's so fast that it can display retouched images in real-time, so that the photographer can see the final version of the image while still framing.

Courtesy of the researchers (edited by MIT News)

This week at Siggraph, the premier digital graphics conference, researchers from MIT's Computer Science and Artificial Intelligence Laboratory and Google are presenting a new system that can automatically retouch images in the style of a professional photographer. It's so energy-efficient, however, that it can run on a cellphone, and it's so fast that it can display retouched images in real-time, so that the photographer can see the final version of the image while still framing the shot.

The same system can also speed up existing image-processing algorithms. In tests involving a new Google algorithm for producing high-dynamic-range images, which capture subtleties of color lost in standard digital images, the new system produced results that were visually indistinguishable from those of the algorithm in about one-tenth the time -- again, fast enough for real-time display.

The system is a machine-learning system, meaning that it learns to perform tasks by analyzing training data; in this case, for each new task it learned, it was trained on thousands of pairs of images, raw and retouched.

The work builds on an earlier project from the MIT researchers, in which a cellphone would send a low-resolution version of an image to a web server. The server would send back a "transform recipe" that could be used to retouch the high-resolution version of the image on the phone, reducing bandwidth consumption.

"Google heard about the work I'd done on the transform recipe," says Michaël Gharbi, an MIT graduate student in electrical engineering and computer science and first author on both papers. "They themselves did a follow-up on that, so we met and merged the two approaches. The idea was to do everything we were doing before but, instead of having to process everything on the cloud, to learn it. And the first goal of learning it was to speed it up."

Short cuts

In the new work, the bulk of the image processing is performed on a low-resolution image, which drastically reduces time and energy consumption. But this introduces a new difficulty, because the color values of the individual pixels in the high-res image have to be inferred from the much coarser output of the machine-learning system.

In the past, researchers have attempted to use machine learning to learn how to "upsample" a low-res image, or increase its resolution by guessing the values of the omitted pixels. During training, the input to the system is a low-res image, and the output is a high-res image. But this doesn't work well in practice; the low-res image just leaves out too much data.

Gharbi and his colleagues -- MIT professor of electrical engineering and computer science Frédo Durand and Jiawen Chen, Jon Barron, and Sam Hasinoff of Google -- address this problem with two clever tricks. The first is that the output of their machine-learning system is not an image; rather, it's a set of simple formulae for modifying the colors of image pixels. During training, the performance of the system is judged according to how well the output formulae, when applied to the original image, approximate the retouched version.

Taking bearings

The second trick is a technique for determining how to apply those formulae to individual pixels in the high-res image. The output of the researchers' system is a three-dimensional grid, 16 by 16 by 8. The 16-by-16 faces of the grid correspond to pixel locations in the source image; the eight layers stacked on top of them correspond to different pixel intensities. Each cell of the grid contains formulae that determine modifications of the color values of the source images.

That means that each cell of one of the grid's 16-by-16 faces has to stand in for thousands of pixels in the high-res image. But suppose that each set of formulae corresponds to a single location at the center of its cell. Then any given high-res pixel falls within a square defined by four sets of formulae.

Roughly speaking, the modification of that pixel's color value is a combination of the formulae at the square's corners, weighted according to distance. A similar weighting occurs in the third dimension of the grid, the one corresponding to pixel intensity.

The researchers trained their system on a data set created by Durand's group and Adobe Systems, the creators of Photoshop. The data set includes 5,000 images, each retouched by five different photographers. They also trained their system on thousands of pairs of images produced by the application of particular image-processing algorithms, such as the one for creating high-dynamic-range (HDR) images. The software for performing each modification takes up about as much space in memory as a single digital photo, so in principle, a cellphone could be equipped to process images in a range of styles.

Finally, the researchers compared their system's performance to that of a machine-learning system that processed images at full resolution rather than low resolution. During processing, the full-res version needed about 12 gigabytes of memory to execute its operations; the researchers' version needed about 100 megabytes, or one-hundredth as much. The full-resolution version of the HDR system took about 10 times as long to produce an image as the original algorithm, or 100 times as long as the researchers' system.

"This technology has the potential to be very useful for real-time image enhancement on mobile platforms," says Barron. "Using machine learning for computational photography is an exciting prospect but is limited by the severe computational and power constraints of mobile phones. This paper may provide us with a way to sidestep these issues and produce new, compelling, real-time photographic experiences without draining your battery or giving you a laggy viewfinder experience."

###

Additional background

PAPER: Deep bilateral learning for real-time image enhancement https://groups.csail.mit.edu/graphics/hdrnet/data/hdrnet.pdf

ARCHIVE: Streamlining mobile image processing http://news.mit.edu/2015/streamlining-mobile-image-processing-1113

ARCHIVE: Removing reflections from photos taken through windows http://news.mit.edu/2015/algorithm-removes-reflections-photos-0511

ARCHIVE: Spruce up your selfie http://news.mit.edu/2014/spruce-your-selfie

Media Contact

Abby Abazorius
abbya@mit.edu
617-253-2709

 @MIT

http://web.mit.edu/newsoffice 

Abby Abazorius | EurekAlert!

More articles from Information Technology:

nachricht Study suggests buried Internet infrastructure at risk as sea levels rise
18.07.2018 | University of Wisconsin-Madison

nachricht Microscopic trampoline may help create networks of quantum computers
17.07.2018 | University of Colorado at Boulder

All articles from Information Technology >>>

The most recent press releases about innovation >>>

Die letzten 5 Focus-News des innovations-reports im Überblick:

Im Focus: First evidence on the source of extragalactic particles

For the first time ever, scientists have determined the cosmic origin of highest-energy neutrinos. A research group led by IceCube scientist Elisa Resconi, spokesperson of the Collaborative Research Center SFB1258 at the Technical University of Munich (TUM), provides an important piece of evidence that the particles detected by the IceCube neutrino telescope at the South Pole originate from a galaxy four billion light-years away from Earth.

To rule out other origins with certainty, the team led by neutrino physicist Elisa Resconi from the Technical University of Munich and multi-wavelength...

Im Focus: Magnetic vortices: Two independent magnetic skyrmion phases discovered in a single material

For the first time a team of researchers have discovered two different phases of magnetic skyrmions in a single material. Physicists of the Technical Universities of Munich and Dresden and the University of Cologne can now better study and understand the properties of these magnetic structures, which are important for both basic research and applications.

Whirlpools are an everyday experience in a bath tub: When the water is drained a circular vortex is formed. Typically, such whirls are rather stable. Similar...

Im Focus: Breaking the bond: To take part or not?

Physicists working with Roland Wester at the University of Innsbruck have investigated if and how chemical reactions can be influenced by targeted vibrational excitation of the reactants. They were able to demonstrate that excitation with a laser beam does not affect the efficiency of a chemical exchange reaction and that the excited molecular group acts only as a spectator in the reaction.

A frequently used reaction in organic chemistry is nucleophilic substitution. It plays, for example, an important role in in the synthesis of new chemical...

Im Focus: New 2D Spectroscopy Methods

Optical spectroscopy allows investigating the energy structure and dynamic properties of complex quantum systems. Researchers from the University of Würzburg present two new approaches of coherent two-dimensional spectroscopy.

"Put an excitation into the system and observe how it evolves." According to physicist Professor Tobias Brixner, this is the credo of optical spectroscopy....

Im Focus: Chemical reactions in the light of ultrashort X-ray pulses from free-electron lasers

Ultra-short, high-intensity X-ray flashes open the door to the foundations of chemical reactions. Free-electron lasers generate these kinds of pulses, but there is a catch: the pulses vary in duration and energy. An international research team has now presented a solution: Using a ring of 16 detectors and a circularly polarized laser beam, they can determine both factors with attosecond accuracy.

Free-electron lasers (FELs) generate extremely short and intense X-ray flashes. Researchers can use these flashes to resolve structures with diameters on the...

All Focus news of the innovation-report >>>

Anzeige

Anzeige

VideoLinks
Industry & Economy
Event News

Leading experts in Diabetes, Metabolism and Biomedical Engineering discuss Precision Medicine

13.07.2018 | Event News

Conference on Laser Polishing – LaP: Fine Tuning for Surfaces

12.07.2018 | Event News

11th European Wood-based Panel Symposium 2018: Meeting point for the wood-based materials industry

03.07.2018 | Event News

 
Latest News

Pollen taxi for bacteria

18.07.2018 | Life Sciences

Biological signalling processes in intelligent materials

18.07.2018 | Life Sciences

Study suggests buried Internet infrastructure at risk as sea levels rise

18.07.2018 | Information Technology

VideoLinks
Science & Research
Overview of more VideoLinks >>>